EvidenceHire
The Second Layer of Verification.
EvidenceHire is not a job board. It sits on top of whatever recruiting pipeline you already have and cross-checks every candidate claim (résumé, LinkedIn, GitHub, portfolios, publications) against real, sourced evidence, producing a cited, ranked, auditable fit report for each candidate against a specific role.
A 7-stage LLM pipeline plus free-text and live AI video-interview assessments, built by AppliPlus on Cloudflare Workers, durable Workflows, and GPT-5. A live example of the production AI engineering we ship for B2B clients.
How It Works
Gather the Evidence
The candidate applies or is imported. EvidenceHire scrapes LinkedIn, fetches every linked source (GitHub, portfolio, Dribbble, Behance, Upwork, publications), and parses the résumé.
Cross-Check Every Claim
A 7-stage LLM pipeline normalizes the profile and checks each skill against all sources (corroborated, weakly corroborated, claim-only, or contradicted) with verbatim citations.
Prove It (Optional)
Assign a free-text or live AI video-interview assessment to fill evidence gaps. No multiple choice: open scenario questions tailored to the candidate and the role.
Get a Ranked Fit Report
A weighted, auditable fit score against the specific job description, plus recruiter intel: outreach templates, interview prep, compensation estimates, and flight-risk.
What No Other Platform Does
Per-Skill Verbatim Citations
Every verdict on a skill points to the exact excerpt and source document, so you can audit the reasoning, not a black-box score.
Multi-Source Cross-Checking
A skill claimed on a résumé but absent from GitHub, LinkedIn, portfolio, and publications is flagged as "claim only"; it never quietly passes the keyword filter.
AI-Manipulation Detection
Assessment answers are checked for AI authorship; the live video interview adds transcript signals plus a vision-based real-person check to flag synthetic candidates.
Free-Text & Live Video Assessments
No MCQs to game. Candidates write open scenario answers or sit a live, conversational AI video interview: a one-time link, no login required.
Flight-Risk Estimation
Tenure patterns, career trajectory, and market demand are analyzed to estimate retention likelihood before you invest in an offer.
Recruiter Intel
Per-candidate outreach templates, interview questions, compensation estimates, and best-approach guidance, generated from the evidence.
Profile Reuse Across Roles
Normalized profiles and verified skills are cached. Re-screening the same talent pool against a new client role costs a fraction of a fresh analysis.
Fair, Bias-Free Scoring
Gender, race, age, religion, disability, and profile photo are explicitly excluded from all scoring. Candidates are ranked only on job-relevant evidence.
The Analysis Pipeline
A durable, cache-aware Cloudflare Workflow. Seven specialized LLM stages turn raw evidence into a verified, cited fit report, and skill assessments feed back in to improve the score.
Normalize
Raw evidence becomes a structured profile (experience, education, publications, claimed skills), every field citing its source.
Skill Evidence Check
Each claimed skill is graded against all sources: corroborated, weakly corroborated, claim-only, or contradicted, with confidence scores.
Leadership Signal
Team size, span of control, budget and P&L ownership, and leadership-style signals, each backed by verbatim examples.
Relocation Fit
Whether the candidate's location and willingness are compatible with the role: strong fit, viable, uphill, or blocked.
Flight Risk
A 0–100 retention estimate with drivers, counterforces, and a time horizon.
Fit Assessment
A weighted composite score against the JD's must-haves: skill match, depth, leadership, relocation, and flight-risk combined.
Recruiter Intel
Actionable outreach, interview prep, and compensation guidance for the recruiter.
Who It's For
Recruitment Agencies & RPOs
Screen one talent pool against many client roles. Cached profiles make re-screening nearly free, and branded PDF shortlists go straight to hiring managers.
High-Volume Corporate Teams
Turn 500 résumés into a ranked, cited shortlist instead of manual triage. Spam-resistant public apply keeps the pipeline clean.
Anti-Résumé-Padding
AI-polished résumés pass keyword filters but not multi-source cross-checking. Unsupported claims are flagged, not ranked.
Fair-Hiring Programs
Objective, evidence-only scoring with protected attributes excluded: defensible, auditable decisions for every candidate.
Plans & pricing
Indicative pricing scaled by screening volume. Book a demo for a quote matched to your hiring needs.
Starter
from $99/mo
For small teams screening a steady trickle of roles.
- ✓Up to 50 candidate analyses / mo
- ✓Cited, ranked fit reports
- ✓Free-text assessments
- ✓Email support
Team
from $399/mo
For agencies and high-volume corporate hiring.
- ✓Up to 500 analyses / mo
- ✓Cached profile reuse across roles
- ✓Live AI video interviews
- ✓Branded PDF shortlists
- ✓Priority support
Enterprise
Contact for pricing
For RPOs and regulated, high-scale programs.
- ✓Unlimited analyses
- ✓SSO & audit logging
- ✓Custom data handling & compliance
- ✓On-prem / dedicated deployment
- ✓Dedicated success manager
Want a verification, assessment, or multi-agent LLM system like this inside your own product? AppliPlus engineers it end to end: talk to us.
Proof
AI outcomes, not demos
Real systems shipped for teams that needed cost savings, speed, and operational leverage.
AI Hiring Verification Platform
Cited, evidence-verified candidate shortlists
Read Case StudyAI Sales Intelligence Platform
3x pipeline without 3x headcount
Read Case StudyAutomating Therapist Booking & Management
80% reduction in admin work
Read Case StudyFrequently Asked Questions
How does EvidenceHire verify candidate claims?
Every claimed skill is cross-checked against real sources (résumé, LinkedIn, GitHub, portfolios, and publications) by a 7-stage LLM pipeline. Each verdict (corroborated, weakly corroborated, claim-only, or contradicted) links to the exact verbatim excerpt and source document, so the reasoning is auditable rather than a black-box score.
Can EvidenceHire detect AI-generated or padded résumés?
Yes. AI-polished résumés pass keyword filters but fail multi-source cross-checking. A skill claimed but absent from every other source is flagged as "claim only" rather than ranked. Assessment answers are checked for AI authorship, and the live video interview adds transcript and vision-based real-person signals to flag synthetic candidates.
Is the scoring fair and bias-free?
Gender, race, age, religion, disability, and profile photo are explicitly excluded from all scoring. Candidates are ranked only on job-relevant evidence against the specific job description, producing defensible, auditable decisions suitable for fair-hiring programs.
Do candidates need to install or sign up for anything?
No. Free-text assessments and the live AI video interview run from a one-time link with no login required. Candidates write open scenario answers or sit a short conversational interview. There are no multiple-choice questions to game.
How is EvidenceHire priced?
EvidenceHire offers Starter, Team, and Enterprise plans scaled by screening volume and features such as cached profile reuse, branded PDF shortlists, and SSO. Indicative pricing is on this page; book a demo to get a quote matched to your hiring volume.
Stop hiring on unverified claims.
Every skill, cross-checked and cited. Every fit score, auditable.
Last updated: July 2026